GPS stopped working, can't figure it out... what else?

Al Johnson openmoko at mazikeen.demon.co.uk
Mon Jul 12 18:23:11 CEST 2010


On Monday 12 July 2010, Yaroslav Halchenko wrote:
> On Sat, 10 Jul 2010, Al Johnson wrote:
> > > * originally GPS worked for me without AGPS at all, just TTFF was
> > > around
> > > 
> > >   2 min
> > 
> > It still should work. TTFF will be ~40s in ideal conditions, longer as
> 
> that is exactly my problem -- it doesn't work at all -- most of the time
> even GPS time is not acquired after tens of minutes being waited.
> 
> just in case -- got account on online assist for ublox, used python
> script from online with my account info and lat/long -- no change.
> 
> is there any more lower level way to troubleshoot GPS (sending/receiving
> expected input etc) than just eye balling stream of outputs from
> gso-gpsd?

fso-gpsd is a level removed from the gps, so isn't much use for debugging. You 
need to look at the debug output from ogpsd, or disable ogpsd entirely and 
look at the serial port directly. I think that's still configured through 
/etc/frameworkd.conf but it's a while since I've updated.
http://wiki.openmoko.org/wiki/OpenmokoFramework#How_to_debug

ogpsd uses the native ubx method to talk to the gps. There have been times in 
the past where duff init data caused the gps to crash, reset and start talking 
NMEA again. In that case you would never see a lock even if the gps had one 
because the ubx parser rejects the NMEA as invalid data. If you enable logging 
you will be able to see all this in the log.



More information about the support mailing list